- Purpose: Set PC(Desktop or Laptop) for genenel perpose use like using word processor, playing game, programming and so on.
1. Change HDD(Hard Disk Drive) to SSD(Solide State Drive) to get faster read/write performance to the main memory.
2. Install two most famous OS (Windows and Ubuntu) because we don't know which OS would be needed in the future.
3. Partition and Format SSD for efficient use in the situation of two OS and one SSD
- Environment & Requirements
Labtop with Intel i5 CPU, 250GB SSD, Windows 10 booting USB Flash Drive, Ubuntu 16.04 LTS booting USB Flash Drive, Windows 10 PC which is used for initializing SSD and making Booting USB Flash Drive
- Prepare SSD
If your SSD was used before at another PC, your SSD need to be clean before use as follows:
1. Connect your SSD to another PC which has it's own main memory and OS on it.
2. Clean your SSD using diskpart program
>diskpart
>list disk
Check SSD disk #. For me, SSD disk # is 1.
>select disk 1
>clean
>convert gpt
>exit
※ GPT is need for UEFI Windows 10.
(No warranty: Google for exact information)
Fig 1. Clean your SSD using 'diskpart' program
※ The 'diskpart' program's 'clean' command use some simple mechanism to clean memory. Thas is, 'clean' command erases memory's fisrt small part and last small part.
This way is possible because the memory usage information is mostly in the first and last part of the memory and without this information this memory is recognized as empty by Users of that memory (e.g. by PC, Printer and so on).
If you want to erase all, use 'clean all' command instead of 'clean'.
(Google for further information)
- Prepare Windows 10 UEFI booting USB
1. Download program called 'MediaCreationTool1809.exe' at link bellow:
3. Follow that program guide to make Windows 10 booting USB Flash Drive
Fig 2. Making Windows 10 booting USB Flash Drive
※ Why need Windows 10 with UEFI booting, not others (e.g. BIOS booting) ?
Ubuntu only supports UEFI booting and I think we need to unify booting mechanism in the situation of one Main Memory (SSD or HDD) and two OS. Need to study further.
(No warranty: Google for exact information)
- Prepare Ubuntu 16.04 LTS booting USB
- Replace HDD with SSD
Open the back cover of a laptop and replace HDD with SSD.
Fig 3. Replacing HDD with SSD (Left: before, Right: after)
- Install UEFI booting Window 10 with the booting USB Flash Drive
1. Enable UEFI boot at your BIOS options.
2. Boot with the booting USB Flash Drive.
3. Run Windows 10 installer and follow guide of the installer (to install Windows 10 on SSD)
4. Boot with SSD to run the newly installed Windows 10.
5. Check if your SSD has partition for EFI. If not, it means that you didn't install Windows 10 with UEFI booting mode and shuold reinstall Windows 10 with UEFI booting mode.
Fig 4. Example of the memory partition which includes EFI partitionFig 5. Warning message during Ubuntu installation because of BIOS booting OS installed on another partition
- Planing for your SSD usage (Partitioning and Formatting) and installing Ubuntu
To use two OS on one Main Memory, you will need two partition for each OS.
You may need shared partition for the two OS which is accessible for the two OS.
For me, partitioned SSD into three region (not counting auxiliary partition like EFI partition, Recovery partition, only counting OS partition, Data partition).
My 250GB SSD was used One for Windows 10 (250*0.4 GB), another for Ubuntu 16.04 LTS (250*0.4 GB), the other for data partition (250*0.2 GB) shared by two OS.
Because we already installed Windows 10 on SSD, follow steps below to partition SSD:
(Stpe 4~5: installing Ubuntu)
1. Turn on laptop to run Windows 10 on SSD.
2. Use windows disk manager and downsize your Windows 10 partition (not EFI partition or others) resulting in 250*0.4 GB for Windows 10 partition
3. Restart laptop and boot with Ubuntu booting USB Flash Drive.
4. Install Ubuntu. During Ubuntu installation, you should select an option to partition by your self.
Fig 6. Option to partition by your self for Ubuntu installation
5. Partion 512 MB for 'EFI system partition' and 250*0.4 GB for Ubuntu.
Fig 7. Example of paritioning for Ubuntu installationFig 8. Result of partitioning for Ubuntu installation (Left: before Ubuntu installation, Right: after Ubuntu installation)
6. Set remaining 250*0.2 GB of SSD as Data partition.
Set remaining 250*0.2 GB as a partition and format the partition as NTFS file system which are accessible for both Window and Ubuntu.
Use burning tools to burn Raspbian OS Image to The SD Card. I used a tool named "Etcher"
Result:
- Booting with Raspbian OS@Raspberry Pi
Insert the SD Card into Raspberry Pi and connect a monitor, mouse and keyboard to the Raspberry Pi(1).
Supply power to the Raspberry Pi by connecting USB Power cable to the Raspberry Pi(2).
Result:
- Notes
(1) Need HDMI cable to connect a monitor to Raspberry Pi.
(2) If your Raspberry Pi are repeating booting process, the reason could be under power. I experienced repeating booting issue when I supply power with 5V/600mA adapter and solved the problem by supplying power with 5V/1A adapter.
TP (True Positive) : Acceptance of the test speaker who is the same with the target speaker.
FP (False Positive) : Acceptance of the test speaker who is not the same with the target speaker.
FN (False Negative) : Rejection of the test speaker who is the same with the target speaker.
TN (True Negative) : Rejection of the test speaker who is not the same with the target speaker.
→ Utterances of an test speaker result in predicted class.
- False acceptance rate (FAR; Miss probability)
The number of verified identities for which the test speaker was different from the target speaker normalized against the total number of acceptances.
In biometrics the instance of a security system incorrectly verifying or identifying an unauthorized person. Also referred to as a type Ⅱ error, a false acceptance typically is considered the most serious of biometric security errors as it gives unauthorized users access to systems that expressly are trying to keep them out.
The number of identities which were not verified for which the test speaker was the same as the target speaker normalized against the total number of rejections.
- Receiver operating characteristic (ROC) curve
In the speaker recognition circles, ROC curve are normally plotted with the FAR on the abscissa and the FRR on the ordinate. In this kind of ROC, the objective is to have the area under the curve approach 0. Generally, a random guess should produce an area of about 0.5.
- Equal-error rate (EER)
The percentage where FRR and FAR are same.
- Detection error trade-off (DET) curve
The DET curve has been favored over traditional ROC curves since its introduction by NIST in 1997. The DET curve plots the miss probability in percentage form versus the false alarm probability, also in percentages. The DET curve always uses a logarithmic scale.
- Detection cost function (DCF)
NIST defines a DCF, , as
where, is the cost of missed detection, is the cost of a false alarm, P(Target)is the a-priori probability of a target speaker, and P(NonTarget)is the a-priori probability of a non-target speaker. NIST uses the following values for the 1999 NIST evaluations,
※ 참고 서적 : Homayoon Geigi, 2011. Fundamentals of Speaker Recognition. Springer Science+Business Media, LLC, New York.
생체 인식 기술을 사용하는 목적에는 identification과 verification이 있다. Identification과 verification은 사용하는 기술과 메커니즘에 있어서 약간의 차이가 있다.
1. Identification(1:n) : ‘Identify’라는 단어는 누군가의 신원을 파악하는 것을 의미한다. 예를 들어, ‘I’m identifying him’은 ‘내가 그를 파악한다’라는 뜻이다. 이 어감에서 알 수 있듯이 사람의 경우 identification은 내가 알고 있는 수많은 사람들(n) 중에 현재 만난 한 사람(1)이 누군지 구별하는 것이다. 마찬가지로 시스템의 경우 identification은 database에 저장된 수 많은 사람들(n) 중에 현재 입력으로 들어온 알려지지 않은 한 사람(1)이 누군지를 찾는 것이다. 따라서 Identification을 위한 시스템은 아래와 같은 특성을 갖는다.
① Database에 충분한 양의 정보를 가지고 있어야 한다. Database에 충분한 정보가 있지 않다면 입력이 들어와도 누군지를 구별하기 힘들다.
②큰 database를 효율적으로 다룰 수 있도록 설계해야 하므로 verification에 비해 더 많은 노력이 필요하다.
③큰 database를 사용하므로 identification을 하는데 많은 처리시간이 요구된다.
Identification 시스템이 사용되는 대표적인 예로 범죄수사기관이 있다. 범죄수사기관은 지문과 얼굴사진과 같은 수많은 개인식별자료를 database화해서 가지고 있다가 범죄현장에서 추출한 용의자의 지문이나 얼굴영상을 database에서 누군지 찾는데 사용한다.
2. Verification (1:1) : ‘Verify’란 어떤 대상물에 대하여 그 대상이 참인가 거짓인가를 확인한다는 뜻이다. 즉, Verification은 한 사람(1)의 정보를 database에 가지고 있다가 현재 입력으로 들어온 알려지지 않은 한 사람(1)이 database에 있는 사람과 일치하는지를 확인하는 것이다.
Identification과 verification을 합쳐서 recognition이라 한다.
